EP0085667A2 - Enzymatic process for the determination of antibiotics containing a beta-lactam nucleus - Google Patents

Abstract

An enzymatic process for the determination of beta-lactam antibiotics in a biological liquid, comprising the steps of: (1) incubating the liquid with soluble D-alanyl-D-alanine-carboxypeptidase produced by Actinomadura R 39, said enzyme being immobilized on a water-insoluble support, the beta-lactam antibiotic reacting with the immobilized enzyme to form an inactive and equimolecular enzyme-antibiotic complex; (2) separating the immobilized enzyme from the liquid and washing it; (3) incubating the immobilized enzyme with a substrate solution to hydrolyze it and form an amount of D-alanine corresponding to the residual enzymatic activity; (4) determining the amount of D-alanine formed; and (5) comparing the determination of step (4) with a standard to obtain the concentration of the antibiotic in the biological liquid; as well as a test set for carrying out this process and comprising the necessary reagents.

Description

The present invention relates to a novel, rapid and sensitive enzymatic method for the determination of beta-lactam nucleus antibiotics in a biological fluid. It also relates to a test kit which can be used for the implementation of this process.

At present, antibiotics are widely used not only as therapeutic agents in the treatment of infectious diseases caused by bacteria, but also as food preservatives and as additives in animal feed to stimulate growth. There is therefore an increasing need to be able to detect the presence of antibiotics, even in very low concentrations, in complex biological fluids, such as milk, urine, blood, serum, saliva, meat extracts, fermentation liquids, etc.

The case of milk production is an example. It is well known in fact to use penicillins to treat certain infectious diseases of milk producing cattle, such as mastitis. However, for obvious medical reasons, milk intended for human consumption must, in principle, be free from all traces of antibiotics. On the other hand, penicillin concentrations of 0.005 IU / ml or less can have detrimental effects in the production of milk-based products, such as cheese, yogurt, etc.

It therefore appears that it is necessary to be able to determine, quickly and precisely, the concentration of penicillins in the milk produced by livestock and this, preferably, directly on site at the farm.

Microbiological procedures have long been used to determine relatively low concentrations of beta-lactam antibiotics in body fluids. These methods are based on measuring the inhibition of the growth of antibiotic-sensitive microorganisms, in the presence of a sample of the biological fluid. However, these processes are very time consuming and highly technical; in the best of cases, the time required to obtain a result is approximately 2 to 3 hours, which is not admissible in practice.

More recently, a rapid microbiological method has also been proposed for detecting the presence of antibiotics in a biological fluid, in particular in milk (cf. American Patent No. 4,239,852).

According to this method, the sample of the liquid to be examined is incubated, on the one hand with cells or parts of cells of a microorganism very sensitive to antibiotics, in particular Bacillus stearothermophilus and on the other hand, with an antibiotic marked with a radioactive element or by an enzyme. During the incubation, the antibiotic possibly present in the sample and the labeled antibiotic compete to bind to the receptor sites of cells or parts of cells. Next, the amount of labeled antibiotic that has been attached to cells or parts of cells is determined; this provides an indication of the presence (or not) of antibiotics, since the amount of labeled antibiotic attached is inversely proportional to the concentration of antibiotic in the sample.

According to the author of this patent, this process would detect antibiotic concentrations as low as 0.01 IU / ml, or even 0.001 IU / ml in milk in just under 15 minutes.

However, the major drawback of this process resides in the fact that, to achieve this result, it is compulsory to have recourse to an antibiotic marked with a radioactive element ( ¹⁴ C or 1251) which must be measured using a special device, such as a scintillation counter. In addition, the handling of radioactive products, even in very small quantities, is not entirely without danger for the person performing the analysis.

It is true that in Example 2 of this patent, a variant of this method is described according to which an antibiotic labeled with an enzyme is used and in which the quantity of labeled antibiotic is determined by a visual colorimetric method. However, this variant can only detect if there is more (or less) than 0.05 IU / ml of penicillin in the milk sample. This process is therefore much less sensitive and therefore much less interesting.

In a recent article, an enzymatic method for determining low concentrations of beta-lactam antibiotics in human serum and in milk is described (JM. FRERE, D. KLEIN and JM. GHUYSEN, Antimicrobial Agents and Chemotherapy, 18, (1980, n ° 4), p.5 ₀ 6-5 ₁₀ ),

This process (designated in the description by "JM. FRERE process") is much more advantageous because it does not require the use of radioactive products requiring sophisticated measuring devices, while being very rapid and of remarkable precision, This process is based on the use of a specific enzyme, in this case the soluble exocellular D-alanyl-D-alanine-carboxypeptidase, which is produced by Actinomadura R 39 (previously called Streptomyces R 39 ). In the description which follows, this enzyme will be designated by "enzyme R 39". As the name suggests, the enzyme R 39 has a specific hydrolysis activity of the D-alanyl-D-alanine end groups of various peptides. By way of example, the _tr ip _e p _t i _of Nalpha N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine undergoes a hydrolysis reaction under the catalytic influence of the enzyme R 39, according to the equation: enzyme

Ac₂-L-Lys-D-Ala + D-alanine.Ac ₂ -L-Lys-D-Ala-D-Ala + H ₂ 0

Ac ₂ -L-Lys-D-Ala + D-alanine.

Another important characteristic of the enzyme R 39 resides in the fact that it reacts with antibiotics with a beta-lactam nucleus to very quickly form an enzyme-antibiotic equimolecular complex, inactive and substantially irreversible.

In the process of JM. FRERE, we take advantage of these properties of the enzyme R 39 to determine very low concentrations of beta-lactam antibiotics. This process has three essential stages. In a first stage, a determined volume of a sample of the liquid to be examined is incubated with a determined quantity of the enzyme R 39. The incubation is carried out under conditions allowing the beta-lactam antibiotic possibly present in the sample react with the enzyme to form an equimolecular enzyme-antibiotic complex, inactive and substantially irreversible.

In a second stage, a determined quantity of substrate, for example alpha N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine, is incubated with the product obtained in the first stage under conditions allowing the hydrolysis of the substrate by the enzyme to form an amount of D-alanine corresponding to the residual enzymatic activity of the enzyme R 39 which was not complexed with the antibiotic in the first stage.

In a third stage, the amount of D-alanine thus formed is determined. Those skilled in the art will readily understand that, depending on the quantity of antibiotic present in the sample, a corresponding quantity of enzyme R 39 will be inactivated at the first stage and that the quantity of D-alanine formed at the second stage (which depends obviously the residual activity of the enzyme) is therefore inversely proportional to the amount of antibiotic present in the sample. If, for example, the sample is free of antibiotics, the R 39 enzyme is not inactivated and an amount of D-alanine is found which corresponds to the total activity of the R 39 enzyme used. artwork. On the other hand, if the sample contains a molar quantity of antibiotic equal to or greater than the molar quantity of enzyme R 39 used, the latter is completely inactivated by the antibiotic and there is no D-alanine formed. Between these two extreme situations, there is an amount of D-alanine which corresponds to the percentage of residual activity of the enzyme R 39. In other words, the amount of D-alanine formed provides an accurate quantitative indication of the antibiotic concentration present in the sample under examination.

In the process of JM. BROTHER, this amount of D-alanine is determined by an enzymatic method. This is based on two coupled enzymatic reactions. In the first reaction, D-alanine is oxidized to pyruvic acid using a D-amino acid oxidase (together with its coenzyme, flavin-adenine-dinucleotide); it is formed in at the same time a corresponding amount of hydrogen peroxide from the oxygen in the air. In the second reaction, the hydrogen peroxide formed is used to oxidize the o-dianisidine using a peroxidase.

This is why, at the third stage of the J-M process. BROTHER, the mixture obtained in the second stage is incubated with a set of reagents comprising a D-amino acid oxidase, its coenzyme (flavin-adenine-dinucleotide; FAD), a peroxidase and o-dianisidine. As the oxidized form of o-dianisidine is colored, at the end of this incubation a brown coloration occurs, the intensity of which depends on the amount of D-alanine.

This therefore makes it possible to determine the amount of D-alanine by a colorimetric method, either visually or by measuring the optical density with a spectrophotometer (at max = 460 nm).

By preparing a series of samples of a known concentration of antibiotic and by applying this method, it is thus possible to establish a calibration curve which relates the percentage of residual enzymatic activity of the enzyme R 39 to the concentration of antibiotic.

To obtain a quantitative indication of the antibiotic concentration of a sample, the procedure is then carried out in exactly the same way and the antibiotic concentration is determined by referring to this calibration curve.

This quantitative evaluation obviously requires the use of a spectrophotometer.

However, in order to determine whether or not the antibiotic concentration exceeds a certain critical value, it is not necessary to use a spectrophotometer or other similar measuring device. It is sufficient to know beforehand at what critical concentration of antibiotic the activity of the enzyme R 39 is completely suppressed or in other words, at what concentration of antibiotic does not form D-alanine and, consequently, it does not produces no coloration at the end of the last incubation. Knowing this critical concentration, it is obvious that it is possible, by a simple visual inspection of the result of the determination, to judge whether or not a given sample contains a concentration of antibiotic less or greater than said critical concentration. We see therefore that by this process it is possible, without having a particular apparatus, to quickly and precisely obtain an indication of the concentration of antibiotic in milk.

In addition, this method makes it possible to determine relatively low concentrations of beta-lactam antibiotic in milk and in human serum. Thus, for example, starting from milk samples with a volume of approximately 20 μl and by incubating those -this with 3 picomoles of enzyme R 39, it is possible to determine, quantitatively (using a spectrophotometer), penicillin concentrations as low as 0.02 IU / ml of milk and, qualitatively, by the visual method described more high, concentrations of more than 0.09 IU / ml milk in less than an hour.

However, the method of J-M. FRERE has various important drawbacks.

Firstly, for the determination of antibiotics in biological fluids such as milk, serum, etc., it is essential to eliminate beforehand from the samples submitted for determination the substances which can disturb the colorimetric determination. In the case of milk, for example, the proteins must first be precipitated with citric acid or rennet, then filter the sample thus treated so as to obtain a clear filtrate which can no longer disturb the colorimetric assay. This protein precipitation operation is complex and tedious, which makes it difficult for the process to be carried out at the place of milk production by unskilled persons.

Second, the sensitivity of the JM process. BROTHER, in particular in the case of biological fluids (milk, saliva, serum, etc.), is insufficient. It is true that this sensitivity could be increased by reducing the amount of R 39 enzyme used. Indeed, if one used for example 0.3 picomole of enzyme instead of 3 picomoles, one could, theoretically, determine antibiotic concentrations which are ten times lower. However, when the quantity of enzyme is reduced, the reaction time between the antibiotic and the enzyme R 39 must also be increased in the same proportions, and the hydrolysis time on the other hand. of the substrate. This means that both the first and the second stage of the process last about ten times longer.

This would remove one of the primary advantages of the J-M process. BROTHER, namely its speed, which is unacceptable.

Furthermore, one could also try to increase the sensitivity, by increasing the volume of the sample which must be incubated with the enzyme R 39. In fact, if a sample with a volume of 200 µl instead of 20 µl, we could theoretically determine antibiotic concentrations which are ten times lower. Furthermore, in this case, it would only be necessary to increase the reaction time between the antibiotic and the enzyme R 39, that is to say only the time of the first stage of the process.

Unfortunately, it is not possible to increase the sensitivity of the JM process in this way. BROTHER, in particular in the case of complex liquids of biological origin. It has in fact been found that it is impossible to carry out a suitable determination when the volume of the sample of biological liquid exceeds a certain critical volume, which varies according to the nature of the biological liquid. Thus, in the case of milk, when the volume of the sample which is subjected to incubation exceeds a volume of ^± 20 μl, it is observed that the amount of oxidized o-dianisidine formed is greatly reduced. The same is true for serum, when the volume of the sample incubated with the enzyme R 39 exceeds - 50 µl.

Finally, in the case of urine, no enzymatic activity is detected, even using sample volumes as small as 10 µl. It is therefore impossible to use this method for the determination of antibiotics in the urine.

It is assumed that body fluids contain substances that inhibit the action of the enzymes used in the process, making it impossible to use large volumes of samples.

This is why, the Applicant has developed a new enzymatic method for the determination of beta-lactam nucleus antibiotics in a biological fluid, which is more improved, in the sense that it does not have the various drawbacks of the methods. state of the art, recalled above.

The Applicant has in fact just discovered that it it is possible to eliminate the various drawbacks of the JM process. BROTHER, by immobilizing the enzyme R 39 on a support insoluble in water, in particular on a poly (N, N-dimethylacrylamide) resin.

• (1) l'incubation dudit liquide avec la D-alanyl-D-alanine-carboxypeptidase soluble produite par Actinomadura R 39, cette incubation étant menée dans des conditions permettant à l'antibiotique bêta-lactame éventuellement présent dans ledit liquide de réagir avec l'enzyme pour former un complexe enzyme-antibiotique équimoléculaire, inactif et sensiblement irréversible;
• (2) l'incubation du produit obtenu à la fin du stade (1) avec une solution de substrat dans des conditions permettant l'hydrolyse dudit substrat par l'enzyme pour former une quantité de D-alanine correspondant à l'activité enzymatique résiduelle;
• (3) la détermination de la quantité de D-alanine formée au stade (2);
• (4) la comparaison de la détermination du stade (3) avec un étalon pour obtenir la concentration de l'antibiotique dans le liquide biologique,

et qui est caractérisé en ce que l'enzyme mise en oeuvre est immobilisée sur un support insoluble dans l'eau et en ce que cette enzyme immobilisée est séparée à partir du liquide biologique et lavée, avant d'être incubée avec la solution de substrat au stade (2).The present invention therefore relates to an improved enzymatic method for the determination of beta-lactam nucleus antibiotics in a biological fluid and comprising the following stages:

• (1) the incubation of said liquid with the soluble D-alanyl-D-alanine-carboxypeptidase produced by Actinomadura R 39, this incubation being carried out under conditions allowing the beta-lactam antibiotic possibly present in said liquid to react with the enzyme to form an equimolecular enzyme-antibiotic complex, inactive and substantially irreversible;
• (2) incubation of the product obtained at the end of stage (1) with a substrate solution under conditions allowing the hydrolysis of said substrate by the enzyme to form an amount of D-alanine corresponding to the residual enzymatic activity ;
• (3) determining the amount of D-alanine formed in step (2);
• (4) comparing the determination of stage (3) with a standard to obtain the concentration of the antibiotic in the biological fluid,

and which is characterized in that the enzyme used is immobilized on a water-insoluble support and in that this immobilized enzyme is separated from the biological liquid and washed, before being incubated with the substrate solution at stage (2).

Unlike the J-M process. FRERE, the method according to the invention makes it possible to make a determination directly on the biological liquid as it is. It is not necessary to remove in advance from the samples, subject to determination, the substances which may disturb the colorimetric determination.

Indeed, as the enzyme R 39 is immobilized on a support insoluble in water, after its reaction with the antibiotic possibly present in the biological liquid, this immobilized enzyme can be easily separated from the biological liquid by a simple filtration and washed . Consequently, during subsequent operations, there is no longer any possibility of liquid interference. biological, because it is already eliminated at the end of stage (1) of the process.

One of the major drawbacks of the J-M process is thus eliminated. BROTHER.

In addition, it was surprisingly found that by applying the method according to the invention, excellent determinations could be made on sample volumes of biological liquid of between 200 μl and 5 ml, without the slightest difficulty.

This is of capital importance. In fact, as explained above, by increasing the volume of the sample which must be incubated with the enzyme R 39, it is possible to increase the sensitivity of the process. One of the drawbacks of the JM process. FRERE lies in the fact that we cannot exceed a critical volume (of the order of 20 µl for milk and 50 _P l for serum). As this drawback is eliminated by the process according to the invention, it is obvious that, unlike that of JM. FRERE, the method of the invention makes it possible to obtain a much greater sensitivity in the determination of biological liquids such as milk, serum, etc. As will be shown later in the exemplary embodiments, the method according to the present invention thus makes it possible, starting from 1 ml milk samples, to determine visually and with certainty, penicillin G concentrations of more than 0.002 IU / ml (1.2 ng / ml) of milk and, using a spectrophotometer, concentrations as low as 0.0005 IU / ml (0.3 ng / ml) of milk in less than one hour. By way of comparison, it should be recalled that the method described in American patent 4,239,852 makes it possible to detect concentrations of the order of 0.01 to 0.001 IU / ml of milk, on the condition however of using radioactive products and a scintillation counter.

In addition, it has been found that the method according to the present invention can be applied successfully, not only for the determination of antibiotics in milk and serum, but also in other complex biological fluids such as for example urine. , blood, saliva, meat extracts, fermentation liquids, etc. Seen from this angle, the process of the invention therefore presents considerable progress compared to that of JM. BROTHER.

The primary advantage of the process of the invention is therefore that it makes it possible to rapidly detect very low concentrations of beta-lactam antibiotic, for example of the order of 0.002 IU / ml, in the most diverse biological fluids, without having to use a special analysis device.

Antibiotics, the concentration of which can be determined using the process according to the invention, belong to the group of antibiotics which are characterized by the presence of a beta-lactam nucleus in their molecule, i.e. principle all penicillins and cephalosporins. By way of example of penicillins, mention may be made of benzylpenicillin (penicillin G), ampicillin, phenoxymethylpenicillin, carbenicillin, methicillin, oxacillin, cloxacillin, etc. By way of example of cephalosporins, mention may be made of cephalosporin C, cefaloglycin, cefalothin, cefalexin, etc., 1

Particularly favorable results have been obtained with penicillin G.

In stage (1) of the process according to the invention, a determined volume of a sample of the biological liquid is incubated with a determined quantity of the enzyme R 39, immobilized on a support insoluble in water.

As explained above, by immobilizing the enzyme R 39 on a support insoluble in water, it is possible to work with very large volumes of sample. By increasing the volume of the sample, the sensitivity of the process is also increased. For example, by doubling the sample volume, the sensitivity is doubled, by tripling the sample volume, the sensitivity is tripled, etc. We can therefore choose the volume of the sample according to the desired sensitivity. In the case of milk, for example, the sensitivity of the process can be adapted to the standards set by the legislation of the country where it is used, or to the requirements of the dairy industry.

It is true that the same effect can also be obtained by reducing the amount of enzyme R 39. However, in this case, it is necessary proportionally increase not only the duration of stage (1), but also the duration of stage (2) of the process, while by increasing the volume of the sample, only the duration of stage (1) of the process is affected. It is therefore preferable to work with a constant quantity of enzyme R 39, for example an amount of approximately 3 picomoles and to adapt the volume of the sample according to the circumstances to the desired sensitivity.

In practice, sample volumes of between 20 μl and 5 ml are used, which allow the determination of very low antibiotic concentrations, for example of the order of 0.002 IU / ml of penicillin, in one hour or less.

The excellent sensitivity, speed and precision of the process according to the invention result from the particular characteristics of the enzyme R 39 on the one hand and, on the other hand, from its immobilization on a support insoluble in water.

• - la formation extrêmement rapide d'un complexe enzyme-antibiotique équimoléculaire inactif,
• - la stabilité extraordinaire dudit complexe parce que, une fois formé, il ne se décompose que fort lentement. A titre d'exemple, le temps de demi-vie du complexe formé entre l'enzyme R 39 et la pénicilline G est d'environ 70 heures à 37°C;
• ― une excellente activité enzymatique se traduisant par une hydrolyse très rapide des groupements terminaux D-alanyl-D-analine du substrat peptidique,

Indeed, the enzyme R 39 is characterized by

• - the extremely rapid formation of an inactive equimolecular enzyme-antibiotic complex,
• - the extraordinary stability of said complex because, once formed, it decomposes only very slowly. For example, the half-life of the complex formed between the enzyme R 39 and penicillin G is approximately 70 hours at 37 ° C;
• - an excellent enzymatic activity resulting in a very rapid hydrolysis of the D-alanyl-D-analine terminal groups of the peptide substrate,

Thanks to these three characteristics, the enzyme R 39 occupies a privileged position compared to the D-alanyl-D-alanine-carboxypeptidda.3es identified so far. Indeed, given that the decomposition time of the enzyme-antibiotic complex is infinitely greater than the total time that it takes respectively for the formation of the complex and for the measurement of the residual enzymatic activity, it is not to fear that the results of the determination may be distorted by release of the active enzyme following premature breakdown of the enzyme-antibiotic complex.

However, when we examine the D-alanyl-D-alanine-carboxypeptidases that we know today, apart from the enzyme R 39, there is no other that perfectly meets these conditions; in fact, it is either the rate of formation of the complex which is ten times slower, or the stability of the antibiotic-enzyme complex which is absolutely insufficient, or else the rate of hydrolysis of the peptide substrate which is much too low (this is particularly the case for all endocellular carboxypeptidases linked to the bacterial membrane),

The enzyme R 39 is the specific soluble exocellular D-alanyl-D-alanine-carboxypeptidase which is excreted by Actinomadura R 39, when this microorganism is cultivated (deposited on July 10, 1981 at the Institut Pasteur in Paris n ° 1- 127) in an appropriate culture medium.

For the implementation of the process of the invention, this enzyme must obviously be substantially pure. Its preparation and its purification can be carried out according to the methods described in the literature (see on this subject the article by JM. FRERE & al., Biochem. J. 143, (1974), 233-240, entitled "Molecular Weight, Aminoacid Composition and Physicochemical Properties of the Exocellular DD-Carboxypeptidase-Transpeptidase of Streptomyces R 39) However, this enzyme, in its pure state, is now commercially available and can be obtained from UCB BIOPRODUCTS, SA (Belgium).

According to the present invention, the soluble enzyme R 39 is immobilized on a support insoluble in water.

The method used for this asset is not critical. For this purpose, it is therefore possible to use all the conventional methods known to those skilled in the art, among which there are more particularly those which proceed either by attachment of the enzyme to the support by a covalent bond, or by physical adsorption of the enzyme on the support, either by trapping the enzyme in a polymer matrix.

Likewise, the nature of the support material is not critical. In principle, any water-insoluble material, organic or inorganic, commonly used for immobilization can be used. tion of an enzyme, As suitable support material, mention may be made of natural polymers, such as cellulose and its derivatives, agarose, starch and its derivatives, dextran, collagen, keratin, etc .; synthetic polymers, such as crosslinked polyacrylamides in the form of pearls or gels; crosslinked polystyrenes; ethylene-maleic anhydride copolymers; polyamides; crosslinked 2-hydroxyethyl polymethacrylates; ion exchange resins, etc., inorganic substances, such as glass, alumina, silica, calcium carbonate, hydroxyapatite, clays, etc.

However, those skilled in the art will understand that it is necessary to choose an immobilization method which leaves all the properties of the enzyme R 39 intact and which in no way affects its enzymatic activity, even after an extended period of storage. .

A particularly preferred support material for carrying out the process according to the present invention is the crosslinked poly (N, N-dimethylacrylamide) resin, the use of which has already been proposed in the synthesis of polypeptides in solid phase (cf. R. ARSHADY et al., J. Chem. Soc.Chem.Commun. 1979, n ° 9.423-425). This resin has the advantage of being easily swellable in a wide variety of organic and aqueous solvents (water, methanol, N, N-dimethylformamide, N, N-dimethylacetamide, pyridine or dichloromethane) and, in addition, to be easy to handle. It is preferably prepared by emulsion copolymerization of a mixture of N, N-dimethylacrylamide, of N, N ^! • ethylenebisacrylamide and the methyl ester of N-acryloylsarcosine.

Under these conditions, a resin is obtained which is in the form of solid particles, for example beads or pearls, the particle size of which is preferably between 0.1 mm to 2 mm, and which can therefore easily be separated by filtration. .

This resin has a three-dimensional crosslinked structure, since the polymer chains are linked together by N, N-ethylenebisacrylamide (which acts as crosslinking agent) and it also carries ester groups, thanks to the introduction of the methyl ester of N-acryloylsarcosine (CH ₂ ⁼ CH-CO-N (CH ₃ ) -CH ₂ -COOCH ₃ ) in the polymer chains.

When the polymerization of this resin is complete, the ester groups carried by this resin are also converted into functional amino groups by reaction with ethylenediamine, according to the equation:

The poly (N, N-dimethylacrylamide) resin thus modified generally contains 0.2 to 1 mmol of ethylenediamine per gram of resin. In this form, it constitutes an ideal support material for the immobilization of the enzyme R 39.

This immobilization is preferably carried out by fixing the enzyme R 39 to the resin by a covalent bond. In fact, unlike other immobilization methods, this method makes it possible to avoid leakage of the enzyme from the support.

The covalent bond of the enzyme R 39 on the poly (N, N-dimethylacrylamide) resin, modified with ethylenediamine as just explained above, can be carried out for example by means of a appropriate coupling containing a functional group capable of reacting with the functional amino groups of the poly (N, N-dimethylacrylamide) resin and another functional group capable of reacting with the free amino groups of the enzyme R 39.

Use is preferably made of coupling agents which comprise ester groups whose half-life time in aqueous medium is sufficiently long, such as for example the N-hydroxysuccinimide diesters of an aliphatic dicarboxylic acid containing 4 to 9 carbon atoms, such as glutaric acid, suberic acid, etc. In fact, as the coupling reaction with the free amino groups of the enzyme R 39 is carried out in an aqueous medium, the functional groups of the coupling agent must be sufficiently stable there.

(n = un nombre entier de 2 à 7)To fix the enzyme R 39 on the poly (N, N-dimethylacrylamide) resin modified with ethylenediamine, this resin is reacted first with a diester of N-hydroxysuccinimide, in a solvent organic suitable (for example N, N-dimétbylacétamide) to avoid the hydrolysis of the ester groups. By this reaction, the functional amino groups of the resin are transformed into functional ester groups, according to the equation:

(n = an integer from 2 to 7)

(n = un nombre entier de 2 à 7)Then, the coupling between the enzyme R 39 and the resin is carried out by reacting the functional ester groups, introduced onto this resin by the ester of N-hydroxysuccinimide, with the free amino groups of the enzyme R 39, according to l 'equation:

(n = an integer from 2 to 7)

The coupling reaction between the enzyme R 39 and the resin is carried out for several hours, at a temperature between 0 and 20 ° C and in an appropriate buffer to adjust the pH to a value between 7 and 8.5, from preferably close to 7.7. It is obvious that this buffer must be free of primary amines capable of to react with the functional ester groups of the resin.

When the coupling reaction is complete, the unreacted ester groups are blocked by the amino groups of compounds such as ethanolamine or the glycine methyl ester. The immobilized enzyme thus obtained is washed with an appropriate buffer and it can be used as it is for carrying out the process according to the present invention.

Using this method, the immobilization yields are excellent: they can vary between 15 and 95%, depending on the pH of the buffer used as well as the duration and the temperature of the coupling reaction.

A detailed example is given below illustrating this method of immobilizing the enzyme R 39 on a poly (N, N-dimethylacrylamide) resin.

Immobilization of the enzyme R 39 on a poly (N, N-dimethylacrylamide) resin. a) Preparation of the -poly resin (N, N-dieethylacryiamide) carrying functional amine groups derived from ethylenediamine.

The poly (N, N-dimethylacrylamide) resin used as a support was prepared by emulsion copolymerization of a mixture of N, N-dimethylacrylamide (12 g), N, N'-ethylenebisacrylamide (1 g) and 1 N-acryloylsarcosine methyl ester (1.4 g), according to the method described in the article by R. ARSHADY et al. (J. Chem. Soc. Chem. Comm. 1979, n ° 9,423-425).

Approximately 12 to 13 g of resin are thus obtained, in the form of beads having a particle size between 0.1 mm and 2 mm.

10 g of this resin are added to 320 ml of ethylenediamine and the mixture obtained is stirred overnight at room temperature.

Then, the resin is filtered and washed successively with N, N-dimethylformamide and water until the filtrate has a neutral pH. Finally, wash with methanol and allow the resin to deflate using diethyl ether.

10 g of resin are thus obtained containing 0.3 mmol of ethylenediamine per gram.

b) Preparation of the poly (N, N-dimethylacrylamide) resin carrying functional ester groups derived from the diester of N-hydroxysuccinimide of glutaric acid. b.1) Preparation of the diester of N-hydroxysuccinimide of glutaric acid.

6.7 g of glutaric acid and 12.7 g of N-hydroxysuccinimide are dissolved in 50 ml of N, N-dimethylacetamide at 0 ° C. A solution of 23.7 g of dicyclohexylcarbodiimide in 50 ml of dichloromethane is added thereto dropwise. The mixture is allowed to return to ambient temperature and, after standing overnight, the reaction mixture is filtered and the filtrate is evaporated. The residue is recrystallized from diethyl ether. The diester of N-hydroxysuccinimide of glutaric acid is obtained with an almost quantitative yield.

b.2) Preparation of the poly (N, N-dimethylacrylamide) resin carrying functional ester groups.

500 mg of poly (N, N-dimethylacrylamide) resin carrying functional amino groups derived from ethylenediamine, prepared as described in a), is suspended in 50 ml of N, N-dimethylacetamide. 2 g of the diester of N-hydroxysuccinimide of glutaric acid (prepared as described in b.l) are added thereto. The mixture obtained is stirred for 24 hours at room temperature. The completion of the reaction is checked by the test by E. KAISER (Anal Biochem. 34, (1970, no2), 595-8). The resin is filtered and washed five times with 100 ml of N, N-dimethylacetamide and five times with 100 ml of methanol, then the resin is allowed to deflate using diethyl ether.

c) Immobilization of the enzyme R 39 on the poly (N, N-dimethylacrylamide) resin carrying functional ester groups derived from the N-hydroxysuccinimide diester of glutaric acid. c.l) Preparation of the enzyme R 39.

The enzyme R 39 used was prepared and purified according to the method described in the article by J-M. FRERE et al. (Biochem. J. 143, (1974), 233-240)

The enzyme R 39 thus purified has a specific activity 19.8 units / mg protein. One unit of R 39 enzyme catalyzes the hydrolysis of 1 µmole of N ^alpha , N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine per minute at 37 ° C, when the enzyme is incubated with a solution 8 mM of the substrate in a 0.03 M Tris-HCl buffer (pH 7.5) containing 3 mM MgCl ₂ .

600 μg of purified enzyme R 39, dissolved in 1 ml of 0.1 M Tris-HCl buffer (pH 7.7) containing 0.2 M NaCl and 0.05 M MgCl _{2 are} subjected to dialysis for 6 hours against 200 ml of 0.1 M Hepes buffer (pH 7.7) containing 0.1 M NaCl and 0.05 M MgCl _2. This operation is repeated four times.

Tris-HCl = 2-amino-2-hydroxymethyl-1,3-propanediol hydrochloride;

Hepes = 4-hydroxyethyl-1-pipererazineethanesulfonic acid.

c.2) Preparation of the resin.

100 mg of poly (N, N-dimethylacrylamide) resin bearing functional ester groups derived from the N-hydroxysuccinimide diester of glutaric acid, prepared as described in b.2) is allowed to swell in 50 ml of N, N-dimethylacetamide. When the beads have reached their maximum swelling point (after about three hours), the resin is filtered and washed five times with 100 ml of 0.1 M Hepes buffer (pH 7.7) and five times with 100 ml of 0.1 M Hepes buffer (pH 7.7) containing 0.1 M NaCl and 0.05 M MgCl ₂

c.3) Coupling between the enzyme R 39 and the resin.

The wet resin, prepared as described in c.2), is placed in a 10 ml flask. 70 μg (1320 picomoles) of enzyme R 39 was added, prepared as described in cl), dissolved in 1.75 ml of 0.1 M Hepes buffer (pH 7.7) containing 0.1 M NaCl and 0.05 M MCl _2. Then, the flask is subjected to a rotational movement of 100 revolutions per minute for 16 hours at room temperature.

The resin is filtered and washed five times with 10 ml of 0.1 M Hepes buffer (pH 7.7) containing 0.1 M NaCl and 0.05 M MgCl ₂ .

After this operation, the resin is subjected to the vacuum of the pump water for 30 minutes,

The weight of the resin thus obtained is 503 mg. The enzymatic activity of the enzyme R 39 immobilized on this resin is 15 ^p g (283 picomoles). The return on the asset is therefore 21%.

The enzyme R 39 immobilized on a solid support has excellent stability and therefore withstands high temperatures up to 70 ° C. This is why the incubation of the biological liquid with the immobilized enzyme R 39 can be carried out in a temperature range between 20 and 50 ° C. Preferably, this temperature is close to 37 ° C. In fact, under these conditions, the duration of the incubation, which is closely linked to the time necessary for the formation of the enzyme-inactive equimolecular antibiotic complex, is approximately 20 minutes. An increase in the temperature of the incubation will have the effect of reducing its duration and vice versa. It is therefore possible to further shorten the duration of the process by increasing the temperature.

In accordance with the present invention, at the end of stage (1) of the process, the immobilized enzyme R 39 is separated from the biological fluid. This separation can be carried out by any suitable method, for example by decantation, by filtration or even by centrifugation. Simple filtration is preferred. The immobilized enzyme is then washed, one or more times, with an appropriate buffer in order to remove the traces of biological fluid which may remain there.

In stage (2) of the process according to the present invention, the immobilized enzyme (freed from the biological fluid) is incubated with a determined quantity of substrate in solution.

During this stage, the fraction of the enzyme which was not consumed in the formation of the enzyme-antibiotic complex in stage (1) of the process, is used to carry out a hydrolysis of a substrate specific for the enzyme R 39. This hydrolysis reaction will therefore produce an amount of D-alanine corresponding to the residual activity of the enzyme R 39.

Given the very nature of the enzyme R 39, it is obvious that it will be possible to use, as substrate, all the peptides which have D-alanyl-D-alanine end groups. By way of non-limiting examples only, let us cite N ^alpha , N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine, N ^alpha -ac'étyl-L-lysyl-D-alanyl-D-alanine , etc.

The amount of substrate used is not critical provided that an amount of substrate is used which is largely in excess of the amount of enzyme R 39 used. In fact, it is always necessary to operate under conditions of saturation of the enzyme by the substrate.

The operating conditions to be observed during this stage are substantially the same as those indicated above for stage (1). The incubation can be carried out in a temperature range between 20 and 50 ° C, preferably at a temperature in the region of 37 ° C. The duration of the incubation must be at least sufficient to obtain a measurable amount of D-alanine with the enzyme R 39 not inactivated. This period is preferably approximately 30 minutes for an incubation temperature of 37 ° C. This duration can be reduced by increasing the temperature of the incubation or, on the contrary, increased by decreasing the temperature of the incubation. It is therefore also advantageous to increase the temperature of this incubation in all cases where the speed of the determination is an important factor.

In order to maintain optimal enzymatic activity, the incubation medium should preferably have a pH value of between 7 and 8.5. Usually a pH value close to 7.7 is maintained by incubating in an appropriate buffer.

In stage (3) of the process according to the present invention, the quantity of D-alanine formed in stage (2) is determined.

This determination of D-alanine can be carried out by any method known per se. It is only important that it is rapid, inexpensive and that it makes it possible to specifically determine D-alanine to the exclusion of all the other products which are present in the liquid to be examined.

This is why the enzymatic methods allowing a colorimetric assay of D-alanine are those which are preferred.

A particularly preferred method is the enzymatic method described by JM. FRERE et al. in "Methods in Enzymology, Vol.XLV, part B, ( ₁₉₇ 6), 610-636"; more particularly on pages 612 and 613.

In this method, D-alanine is oxidized to pyruvic acid using a D-amino acid oxidase (together with its coenzyme, flavin-adenine-dinuelotide); an amount of hydrogen peroxide is formed at the same time corresponding to the amount of D-alanine. This hydrogen peroxide is in turn used to oxidize o-dianisidine using a peroxidase. As the amount of oxidized o-dianisidine thus formed depends closely on the amount of D-alanine and as the oxidized o-dianisidine produces coloration, the measurement of the intensity of the color thus produced can be used for the determination of the amount of D-alanine by a colorimetric assay method.

It is obvious that this system is essentially based on the use of a substance, in this case o-dianisidine, whose oxidized form and reduced form have different colors. Such substances are known in the art as "redox indicators".

There are currently a large number of redox indicators which can usefully replace o-dianisidine in the aforementioned method. By way of example, let us cite the ammonium salt of 2,2'-azinobis- (3-ethyl-2,3-dihydro-6-benzothiazolesulfonic acid), 4-aminoantipyrine in the presence of phenol or N, N-dimethylaniline, etc.

This is why, according to a preferred embodiment, the amount of D-alanine is determined by incubation of the mixture obtained in stage (2) with on the one hand a D-amino acid oxidase, which catalyzes the oxidation of D -alanine to pyruvic acid (with simultaneous formation of hydrogen peroxide) and on the other hand a peroxidase and a redox indicator, which is oxidized by the hydrogen peroxide formed, by means of peroxidase, to produce a coloration of which the intensity is a function of the amount of D-alanine.

In conjunction with D-amino acid oxidase, its coenzyme, flavin-adenine-dinucleotide (FAD), is still used.

In this preferred embodiment, it is possible in principle use any redox indicator capable of being oxidized by hydrogen peroxide, in the presence of peroxidase as catalyst. Particularly preferred redox indicators are o-dianisidine, the oxidized form of which has a brown coloration (maximum absorption at λ = 460 nm) and the ammonium salt of 2,2'-azinobis acid (3- ethyl-2,3-dihydro-6-benzothiazolesulfonique), whose oxidized form has a green color (maximum absorption at λ = 420 nm).

It will be noted that the brown coloration of the oxidized o-dianisidine can advantageously be changed to a bright pink coloration by the addition of sulfuric acid. This pink coloration is very stable, which is very important when it is desired to maintain the result of the determination for a certain time.

In addition, in order to ensure rapid determination, it is possible to execute stages (2) and (3) simultaneously. In this embodiment, the reagents allowing the determination of D-alanine are added directly to stage (2) of the process to the mixture of the immobilized enzyme and the substrate solution and a single incubation of the whole is carried out, under conditions substantially identical to those indicated above for stage (2).

In stage (4) of the method, the determination of stage (3) is compared with a standard to obtain the concentration of the antibiotic in the biological fluid.

The quantitative determination of the antibiotic concentration can be carried out according to the following method.

First, a series of biological fluid samples are prepared containing a known concentration of beta-lactam antibiotic.

This series will contain, in addition to a number of samples containing an increasing concentration of antibiotic, two samples of the biological fluid free of antibiotic.

Then, all of these samples are treated in a strictly identical manner by following steps (1), (2) and (3) of the process according to the present invention.

For one of the antibiotic-free samples, replace however, the substrate solution used in step (2) with an identical volume of water. In this particular case, there will therefore be obtained, at the end of stage (3), a solution which contains all the reagents except the substrate. Given the absence of said substrate, D-alanine will therefore not form and consequently o-dianisidine will not be oxidized. The solution obtained will have a light yellow coloration. This sample will hereinafter be referred to as a "blank sample".

The other antibiotic-free sample, on the contrary, produces a pronounced brown coloration. In fact, as the sample is free of antibiotics, the enzyme R 39 is not inactivated in stage (1) and a maximum amount of D-alanine will be formed (corresponding to the total activity of the enzyme R 39 implementation) and therefore a maximum amount of oxidized o-dianisidine and consequently there will be a pronounced brown coloration. This sample will be designated hereinafter by "control sample".

It will also be understood that when the samples contain a molar quantity of antibiotic which is less than the molar quantity of enzyme R 39 used, only a fraction of this enzyme will be inactivated by the antibiotic in stage (1); in these cases, an amount of D-alanine will therefore form which will correspond to the residual activity of the fraction of the enzyme which has not been inactivated by the antibiotic and therefore also a corresponding amount of o- oxidized dianisidine. In these cases, therefore, there will also be a brown coloration, but of a lower intensity than that observed with the control sample.

Finally, when the samples contain a molar quantity of antibiotic which is equal to or greater than the molar quantity of enzyme R 39 used, the enzyme will be completely inactivated by the antibiotic during stage (1); D-alanine will therefore not form and o-dianisidine will not be oxidized. In these cases, a light yellow coloration identical to that observed with the white sample will therefore be obtained.

To obtain a precise determination, the optical density of the colorings obtained with all the samples, including the blank and control samples, is measured with a spectrophotometer.

As there is also a certain value of optical density for the white sample, it is necessary to deduce this value from those found respectively for the control sample and for the other samples.

From the optical density values thus obtained, the percentage of residual activity (of the enzyme R 39) is calculated for each sample. This percentage is equal to the ratio, multiplied by 100, between the optical density values found respectively for a given sample and for the control sample.

A graph is then drawn, plotting on the abscissa, the antibiotic concentrations and, on the ordinate, the percentages of residual activity of the enzyme R 39.

A straight line is obtained which respectively intersects the ordinate at a point corresponding to the control sample (100% residual enzymatic activity) and the abscissa at a point corresponding to the sample containing an amount of antibiotic equal to the amount molar of enzyme R 39 used (0% residual enzymatic activity).

The graph thus obtained constitutes a "standard" which makes it possible to determine an unknown concentration of beta-lactam antibiotic in a sample of the biological fluid which served to establish it. For this purpose, this sample is treated in a strictly identical manner by following stages (1), (2) and (3) of the process according to the invention; the optical density of the coloration obtained is measured with a spectrophotometer; the value of the optical density found for the blank sample is subtracted therefrom; the percentage of residual enzymatic activity is calculated as indicated above and the antibiotic concentration of the sample is obtained by referring to the standard.

It is thus possible to quantitatively determine antibiotic concentrations as low as 0.0005 IU / ml of biological fluid in a little less than an hour.

However, this method requires the use of a spectrophotometer and must therefore, in principle, be carried out in a laboratory. However, if we only consider whether the antibiotic concentration exceeds a certain threshold or not (for example a concentration maximum imposed by the dairy industry), it is not necessary to use a spectrophotometer. However, this requires some explanation beforehand.

As seen above, the calibration curve cuts the abscissa at a point corresponding to a sample containing an amount of antibiotic equal to the molar amount of enzyme R 39 used. At this critical concentration, the percentage of residual enzymatic activity is zero%, because at the end of stage (3) of the process according to the invention, a light yellow coloration identical to that obtained with the white sample is obtained.

Above this critical concentration, a light yellow coloration is also obtained, because the percentage of residual enzymatic activity is always zero%. On the other hand, below this critical concentration, a brown coloration is obtained because a certain percentage of residual enzymatic activity remains.

Consequently, based simply on the coloration observed at the end of stage (3) of the process, it can be evaluated directly in a sample whether the antibiotic concentration exceeds or does not exceed said critical concentration.

It is therefore sufficient to know said critical concentration, so that one can then, without resorting to a spectrophotometer, quickly determine whether a sample contains a concentration of beta lactam antibiotic which exceeds (or not) this critical concentration. To this end, this sample is treated in an identical manner according to stages (1), (2) and (3) of the process according to the invention and then simply observing the coloration obtained; if the latter is light yellow, the antibiotic concentration will be at least equal to the critical concentration; if, on the other hand, it is brown, the antibiotic concentration will be lower than the critical concentration.

It is thus possible to determine, visually and with certainty, whether the samples contain more or less than 0.002 IU of antibiotic per ml of biological fluid and this in a time of less than one hour.

This method is therefore perfectly suited for the serial examination of milk samples, even outside a laboratory, for example on site at the farm, by non-specialized personnel.

The method for quantitative and qualitative determination of the antibiotic concentration in accordance with the present invention has just been explained in detail with particular reference to the color change produced by o-dianisidine. However, a person skilled in the art will readily understand that, by using other redox indicators in this method, only the color observed will be different.

Another object of the present invention is to provide a test kit which can be used for carrying out the method according to the invention, that is to say which can be used for the determination of a beta-lactam antibiotic in a biological liquid.

• (1) une quantité déterminée de D-alanyl-D-alanine-carboxypeptidase soluble produite par Actinomadura R 39, ladite enzyme étant immobilisée sur un support insoluble dans l'eau;
• (2) une quantité déterminée de substrat;
• (3) les réactifs permettant la détermination de la D-alanine; et
• (4) éventuellement un étalon, par rapport auquel les résultats d'essais réalisés avec les réactifs (1), (2) et (3) peuvent être comparés.

This kit includes

• (1) a determined quantity of soluble D-alanyl-D-alanine-carboxypeptidase produced by Actinomadura R 39, said enzyme being immobilized on a support insoluble in water;
• (2) a determined quantity of substrate;
• (3) reagents for the determination of D-alanine; and
• (4) optionally a standard, against which the results of tests carried out with the reagents (1), (2) and (3) can be compared.

According to a preferred embodiment, the enzyme R 39 is immobilized on a crosslinked poly (N, N-dimethylacrylamide) resin, the substrate is the tripeptide Nal ^pha N ^epsilon -diacetyl-L-lysyl-D-alanyl-D- alanine, while the reagents consist of (a) a D-amino acid oxidase (together with its coenzyme, flavin-adenine-dinucleotide), (b) a peroxidase and (c) a redox indicator, for example o -dianisidine.

As a standard, it is possible optionally to incorporate into the kit a graph establishing the relationship between the antibiotic concentrations and the percentages of residual activity of the enzyme R 39. It is thus possible to carry out quantitative determinations as has been explained above. However, this graph is not essential. While the kit must be used only to determine if the antibiotic concentration exceeds a certain limit value, simply indicate in the instructions for what antibiotic concentrations a turn ^of the observed coloring, after treatment of the sample according to the process according to the invention.

According to a particularly advantageous embodiment, the immobilized enzyme is placed in a syringe provided at its bottom with a plate of filter material, for example sintered glass. Indeed, this device facilitates the separation by filtration of the enzyme R 39 immobilized from the biological liquid, as well as its washing, at the end of stage (1) of the process according to the invention.

The following examples illustrate the present invention without, however, limiting it.

Example 1.

• (a) Préparation des réactifs.
  • 1) Enzyme R 39 immobilisée. L'enzyme R 39 a été immobilisée sur une résine de poly(N,N-diméthyl- acrylamide) suivant la méthode décrite dans l'exemple donné précédemment dans la description.
  • 2) Solution de substrat. On prépare une solution contenant 6 _mg de _N ^alpha N^epsilon-diacétyl-L-lysyl-D-alanyl-D-alanine dans 1 ml d'eau.
  • 3) Suspension de D-aminoacide-oxydase. On prépare une suspension contenant 5 mg de D-aminoacide-oxydase (activité spécifique: 15 U/mg) par ml d'une solution aqueuse 3 molaire de sulfate d'ammonium.
  • 4) Solution de flavine-adénine-dinucléotide (FAD). On prépare une solution contenant 500 µg de FAD dans 6 ml de tampon Tris-HCl 0,1 M (pH : 8,0).
  • 5) Solution de peroxydase. On prépare une solution contenant 50 µg de peroxydase par ml d'eau.
  • 6) Solution d'o-dianisidine. On prépare une solution contenant 2,6 mg de dichlorhydrate d'o-dianisidine par 500 µl d'eau.
• (b) Mode opératoire,

This example illustrates the application of the method according to the present invention for the determination of low concentrations of penicillin G in milk,

• (a) Preparation of the reagents.
  • 1) Immobilized R 39 enzyme. The enzyme R 39 was immobilized on a poly (N, N-dimethylacrylamide) resin according to the method described in the example given previously in the description.
  • 2) Substrate solution. A solution containing 6 _m g of _N ^alpha N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine in 1 ml of water is prepared.
  • 3) Suspension of D-amino acid oxidase. A suspension is prepared containing 5 mg of D-amino acid oxidase (specific activity: 15 U / mg) per ml of a 3 molar aqueous solution of ammonium sulfate.
  • 4) Flavin-adenine-dinucleotide solution (FAD). A solution is prepared containing 500 μg of FAD in 6 ml of 0.1 M Tris-HCl buffer (pH: 8.0).
  • 5) Peroxidase solution. A solution is prepared containing 50 μg of peroxidase per ml of water.
  • 6) O-dianisidine solution. A solution is prepared containing 2.6 mg of o-dianisidine dihydrochloride per 500 μl of water.
• (b) Procedure,

A series of samples of 1 ml of milk, each containing a known concentration of penicillin G, is prepared, as well as two samples (white and control) of 1 ml of milk, free of penicillin G.

To each sample, 5.5 mg of immobilized enzyme R 39 (= 3 picomoles of enzyme) are added and the mixture is incubated for 20 minutes at 37 ° _C.

The milk is then filtered off from the immobilized enzyme and the immobilized enzyme is washed three times with 1 ml of 0.1 M Hepes buffer (pH = 7.7) containing 0.1 M NaCl and MgCl ₂ 0.05 M.

Then, to the immobilized enzyme thus separated and washed, 20 μl of 0.1 M Hepes buffer (pH = 7.7) containing 0.1 M NaCl and MgC ^l l ₂ 0.05 M and ie 10 μl are added. substrate solution (in the case of the control sample and samples containing penicillin G), i.e. 20 μl of the Hepes buffer and 10 μl of water (in the case of the white sample). The mixture is incubated for 30 minutes at 37 ° C.

At the end of this incubation, 2 μl of the suspension of D-amino acid oxidase, 60 μl of the FAD solution, 10 μl of the peroxidase solution and 5 μl of the o-dianisidine solution are added to the product obtained respectively. ; then the mixture is incubated again for 10 minutes at 37 ° C.

The optical density of the colored solutions obtained is then measured with a spectrophotometer. To this end, 1 ml of a methanol-water solution (proportions 1: 1) is added to the cuvette and the optical density is measured at 460 nm.

The optical density value found for the white sample is deduced from those found respectively for the control sample and for the other samples. Finally, the percentage of residual enzymatic activity for each sample is calculated from the optical density values thus obtained.

The results obtained with two series of samples of 1 ml of milk containing variable concentrations of penicillin G are reproduced in table 1

The results thus obtained are reproduced in the form of a graph, shown in the accompanying drawing. This graph shows the penicillin G concentrations in IU / ml on the abscissa and the percentages (symbol%) on the ordinate of residual enzymatic activity.

This graph constitutes a standard which makes it possible to determine the concentration of penicillin G in other samples of 1 ml of milk. For this purpose, these samples are treated in the manner described in this example and their respective concentration of penicillin G is determined by referring to the standard.

It appears from Table I that for a sample containing 0.002 IU of penicillin or more per ml of milk, there is a residual enzymatic activity of the order of zero% and a light yellow coloration identical to that of the white sample is observed. On the other hand, for samples containing less than 0.002 IU of penicillin per ml of milk, a certain percentage of residual enzymatic activity is found and a yellow brown or brown coloration is observed.

Consequently, based simply on the coloration observed at the end of the last incubation, it can be determined, visually and with certainty, whether a milk sample contains a concentration at least equal to 0.002 IU of penicillin per ml of milk, ie 1, 2 ng / ml milk. Indeed, at this concentration or at a higher concentration, a light yellow coloration identical to that of the white sample is obtained. On the other hand, below this concentration (for example if the sample is free of penicillin), the color changes from light yellow to brown:

It can therefore be seen that the process according to the invention makes it possible to very quickly carry out a determination of very low concentrations of penicillin in milk without having to go back to a spectrophotometer or to another similar analysis device. This shows the advantage of this process, which makes it possible to carry out rapid and serial determinations in milk on the very place of its collection, that is to say on the farm.

In addition, it is obvious that this method can also be used to carry out quantitative determinations in the laboratory. Indeed, the graph represented on the accompanying drawing clearly shows that it is possible, using a spectrophotometer, to determine concentrations as low as 0.0005 IU of penicillin per ml of milk, ie 0.3 ng / ml of milk.

Example 2.

This example illustrates the application of the method according to the present invention, for the determination of low concentrations of penicillin G in serum and in urine.

The procedure is as in Example 1 and using the same reagents, but the samples of 1 ml of milk are replaced by samples 1 ml of serum or 1 ml of urine.

The results obtained are reproduced in Tables II and III, respectively for serum and urine.

Tables II and III clearly show that the method according to the invention can be successfully applied to visually determine whether a serum or urine sample contains a concentration at least equal to 0.003 IU, penicillin per ml of serum or d 'urine.

Indeed, at this concentration or at a higher concentration, the observed coloration will be light yellow.

It should be noted that, in the case of urine, it is impossible to make such a determination by the method of J-M. FRERE, in which the soluble but not immobilized enzyme R 39 is used, even on urine volumes as small as 10 μl.

Example 3.

This example illustrates the application of the method according to the present invention to the determination of cephalosporin C in serum.

The procedure is as in Example 1 and using the same reagents, but the samples of 1 ml of milk are replaced by samples of 1 ml of serum each containing a known concentration of cephalosporin C.

The results obtained are reproduced in the following Table IV:

Example ₄ .

This example illustrates the application of the method according to the present invention to the determination of cephalosporin C in serum and carried out at a higher temperature.

The procedure is as in Example 3, but for each incubation, the operation is carried out at a temperature of 47 ° C. and the duration of each incubation is halved.

In addition, to each 1 ml serum sample, 100 μl of 0.5 M Hepes buffer (pH 7.7) containing 0.5 M NaCl and 0.25 M MCl ₂ is added beforehand.

The results obtained are reproduced in Table V below:

This example clearly shows that by increasing the incubation temperature it is possible to significantly decrease the duration of the process.

Claims (12)

1. Enzymatic method for the determination of beta-lactam nucleus antibiotics in a biological fluid comprising the following stages: (1) the incubation of said liquid with the soluble D-alanyl-D-alanine-carboxypeptidase produced by Actinomadura R 39, this incubation being carried out under conditions allowing the beta-lactam antibiotic possibly present in said liquid to react with the enzyme to form an equimolecular enzyme-antibiotic complex, inactive and substantially irreversible; (2) incubation of the product obtained at the end of stage (1) with a substrate solution under conditions allowing the hydrolysis of said substrate by the enzyme to form an amount of D-alanine corresponding to the residual enzymatic activity ; (3) determining the amount of D-alanine formed in step (2); (4) comparing the determination of stage (3) with a standard to obtain the concentration of the antibiotic in the biological fluid,
characterized in that the enzyme used is immobilized on a water-insoluble support and in that this immobilized enzyme is separated from the biological liquid and washed, before being incubated with the substrate solution in stage ( 2). 2. Method according to claim 1, characterized in that the water-insoluble support is a crosslinked poly (N, N-dimethylacrylamide) resin. 3. Method according to claim 2, characterized in that the immobilization is carried out by fixing the enzyme on the poly- (N, N-dimethylacrylamide) resin by a covalent bond. 4. Method according to any one of claims 1 to 3, characterized in that the stages (2) and (3) are executed simultaneously. 5. Method according to any one of claims 1 to 4, characterized in that the biological liquid is chosen from milk, serum, urine, blood, saliva, meat extracts and fermentation fluids. 6. Method according to any one of claims 1 to 5, characterized in that the antibiotic to be determined is chosen from benzylpenicillin, ampicillin, phenoxymethylpenicillin, carbenicillin, methicillin, oxacillin, cloxacillin, cephalosporin C, cefaloglycin, cefalothin and cefalexin. 7. Method according to any one of claims 1 to 6, characterized in that the substrate is chosen from N ^alpha , N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine or N ^alpha -acetyl- L-lysyl-D-alanyl-D-alanine. 8. Method according to any one of claims 1 to 7, characterized in that the amount of.D-alanine is determined by incubation of the mixture obtained in step (2) with, on the one hand, a D-amino acid oxidase , which catalyzes the oxidation of D-alanine to pyruvic acid, with simultaneous formation of hydrogen peroxide and, on the other hand, a peroxidase and a redox indicator, which is oxidized by the hydrogen peroxide formed, by means peroxidase, to produce a color whose intensity is a function of the amount of D-alanine. 9. Test kit for the determination of beta-lactam antibiotics in a biological fluid comprising: (1) a determined quantity of soluble D-alanyl-D-alanine-carboxypeptidase produced by Actinomadura R 39, said enzyme being immobilized on a support insoluble in water; (2) a determined quantity of substrate; (3) reagents for the determination of D-alanine; and (4) possibly a standard, against which the results of tests carried out with the reagents (1), (2) and (3) can be compared. 10. Test kit according to claim 9, characterized in that the water-insoluble support is a crosslinked poly (N, N-dimethylacrylamamide) resin. 11. Test kit according to any one of claims 9 and 10, characterized in that the substrate is chosen from N ^alpha , N ^epsilon -diacetyl-L-lysyl-D-alanyl-D-alanine or N ^alpha -acetyl-L-lysyl-D-alanyl-D-alanine. 12. Test kit according to any one of claims 9 to 11, characterized in that the reagents (3) comprise (a) a D-amino acid oxidase, (b) a peroxidase and (c) a redox indicator.

Images